A Method for Predicting Ultra-Deep Ground Stress in the Ordovician No. 6 Fault Zone of Shunbei Oilfield Based on Coupled Crack Volume and Pore Pressure Parameters with Weakened Nonlinear Higher-Order Disturbances

The Ordovician No. 6 fault zone reservoir in the Shunbei oilfield exhibits characteristics of ultra-deep burial and high-pressure, high-temperature conditions, with pronounced structural control and significant heterogeneity. This renders traditional in-situ stress prediction methods—based on linear elasticity and anisotropy assumptions—inadequate for accurately characterizing the evolution of carbonate reservoir stiffness and its associated uncertainties. To address this, this study integrates wave impedance inversion with high-confining-pressure granular flow bi-axial tests to establish a common-origin constitutive calibration system for seismic and experimental data. Simultaneously, it introduces fracture volume coefficients and pore pressure coefficients to develop a physically interpretable nonlinear weakening function. This function dynamically corrects elastic coefficients, thereby quantitatively characterizing the nonlinear evolution of reservoir stiffness under coupled fracture-pore pressure interactions. The model incorporates a higher-order derivative constraint mechanism, significantly enhancing response continuity and numerical stability in high-gradient disturbance zones and stiffness discontinuity regions. This effectively suppresses numerical oscillations and divergences commonly observed in complex fracture-dissolution systems. Results demonstrate high concordance between predicted in-situ stresses and measured logging data at three key wells, consistent with regional stress evolution patterns. Sensitivity analysis indicates Young's modulus goodness-of-fit improved from 0.89 to 0.95, mean squared error reduced by 43%, and outlier proportion decreased below 1%, confirming the model's effective control over stiffness discontinuities and numerical instability in high-gradient zones. Overall, this study provides a novel methodology for predicting in-situ stresses in ultra-deep carbonate reservoirs, offering engineering guidance and parameterisation references for scheme deployment in complex fractured karst systems.